The mobile telephone industry has been associated with tremendous growth over the last several years. For instance, in the recent past, mobile telephones were only available to those of highest economic status due to service costs and costs associated with mobile phones. Moreover, network coverage was not extensive enough to enable robust service. In particular, only areas associated with dense population were provided with extensive wireless network coverage. Still further, the mobile phones that could utilize the networks to communicate were quite bulky, causing portation of the phone over any significant distance to be difficult at best. In more detail, antennas associated with these phones could be over a foot in length, thus making it difficult to utilize the phones in automobiles or other congested areas.
In contrast, today's portable phones (and other portable devices) can be utilized as full-service computing mechanisms. For example, many of the most recent and advanced mobile phones can be associated with word processing software, accounting software, and various other types of software. Furthermore, network coverage has expanded to cover millions, if not billions, of users. Additionally, mobile phones have decreased in both size and cost. Specifically, modern mobile phones are often small enough to slip into an individual's pocket without discomforting the individual. Furthermore, many mobile network service providers offer phones at extremely low cost to customers who contract for service with such providers.
The increase in the number of mobile users has yielded increased network traffic associated with wireless telecommunication carriers' backend services and/or systems. By way of illustration, backend services, such as a billing system, can be provided by a wireless telecommunication carrier to various clients including customer care call centers, retail stores, payment systems, etc. Typically, the clients can communiate with the backend via a middle tier that can broker data transmission between the clients and the backend system(s) and/or service(s). However, use of such a middle tier can slow network traffic and/or cause failure of the entire system due at least in part to limited numbers of middle tier resources.
For example, a middleware tier can receive and/or transmit requests for several backend services from disparate backends. Oftentimes, services from one or more backends can become slow and/or unresponsive which can cause middleware resources to be occupied until the backend services recover. During this time, requests for responsive backend services can be starved of middleware resources. By way of illustration, all middleware resources can be allocated to unresponsive requests and request queues can fill, which can cause middleware applications to become completely unavailable for further processing.
Services can be associated with disparate markets such that each of the disparate markets can utilize physical instances of the backend system(s) and/or service(s). Traffic can generally flow from a front end system to a backend related to a particular market and then return to the front end, or vice versa. Backend billing system outages, which can be limited to a single market, can produce middleware resource starvation and catastrophic failure, often within five to ten minutes of the backend outage. Conventional techniques that utilize rolling bounces to clear the congestion can yield loss of all requests and oftentimes only temporarily purged the backlog.